Why you might need to identify a coordinate format
The most common coordinate problem is not converting numbers. It is receiving numbers with no explanation of what they are. A CSV from a contractor, a CAD drawing with no projection set, a pair of values copied out of an old report, a survey point someone read out over the phone. You have two large numbers and nothing else, and every tool you open asks you to choose the coordinate system before it will do anything.
This tool answers the question that comes before conversion: what are these numbers? It works out the format, and when the format alone is not enough, it tests the supported systems and zones the numbers could belong to and shows you which ones hold up.
- Unlabelled survey or GIS data: eastings and northings with no zone, no units and no metadata.
- Mystery CAD coordinates: a drawing that lands in the wrong place because nobody knows whether it was State Plane feet, UTM metres or a local grid.
- Latitude and longitude in the wrong order: the single most common silent error in coordinate handling, and one this tool tests explicitly.
- Checking a guess: you suspect a pair is UTM zone 30N but want to see where it would land in the neighbouring zones before committing.
How the coordinate format identifier works
There are two separate questions and the tool treats them separately. The first is format detection: is this decimal degrees, degrees and minutes, degrees minutes seconds, an MGRS reference, a UTM string with a zone, a UK grid reference with two letters, or a bare pair of projected grid numbers? Some formats spell out their own system in their syntax. An MGRS square, a UTM zone number with a hemisphere letter, or the two-letter prefix of a British grid reference each pin the system down completely. When the format is unambiguous the tool confirms it and stops.
The second question only arises for a bare pair of large numbers. Format detection can say "this is some projected grid" but not which one, and that is the part that matters. So the tool goes further with a plausibility check:
- 1. Build the candidate list. British National Grid, all 60 UTM zones in both hemispheres, and every US State Plane zone in both survey feet and metres. Both value orders are tried unless your input labels them.
- 2. Prune with your hint first. If you say the data is from Texas, only Texas State Plane zones and the UTM zones that cover Texas are tested. This is the main lever for both speed and confidence, which is why the hint is applied before scoring, not after.
- 3. Convert every survivor to latitude and longitude using the same conversion code the rest of CoordinateMapper runs on.
- 4. Score each result. Does the point fall inside the area that system covers, the extent of the UK grid or the approximate extent of a State Plane zone? For State Plane, does the nearest US county centre belong to that same zone? Does it land on land or in open ocean? Does it fall inside the region you hinted at?
- 5. Rank and explain. Interpretations that resolve somewhere impossible are discarded or pushed to the bottom. The top two to four are shown with the reason for their ranking, never a bare percentage.
UTM gets special treatment. A UTM easting and northing look identical in every zone, so without a hint the same pair produces up to 60 valid points at the same latitude. Rather than showing 60 cards, the tool groups them into one and lets you pick the zone you expect. Add a region hint and the group usually collapses to one or two.
How accurate is coordinate format detection?
Results here are best guesses to verify against your source, and the tool is deliberately upfront about that. A pair of grid numbers with no metadata does not contain enough information to be certain, and any tool that claims otherwise is hiding the uncertainty rather than removing it. The transparency is the point: you see every interpretation that survived, where each one lands, and why it ranked where it did.
- Strong match means the point sits inside the system's coverage area, on land, and matches your region hint if you gave one. It is the interpretation to try first.
- Plausible means it passes most checks but not all, for example on land but with no hint to corroborate it, or the values had to be read in reverse order.
- Unlikely means it is mathematically valid but resolves somewhere implausible, typically open ocean, outside the system's coverage, or in the territory of a neighbouring State Plane zone. It is shown so you can see it was considered.
The land check is coarse by design, using continental outlines rather than a detailed coastline, so a genuine coastal or offshore coordinate is never wrongly rejected. It cannot tell you whether a point is on the correct street; it can tell you whether an interpretation puts your survey in the middle of the Atlantic. The datum is a separate question the numbers cannot answer either: a bare UTM pair could be WGS84 or NAD83 (sub-metre difference) or NAD27 (tens of metres). The converter lets you change it once you have picked a system.
Why a "Strong match" sometimes lands in the sea. The land check uses coarse continental outlines, not a detailed coastline. If a point falls within roughly 10 to 20 km of a continental landmass, the check treats it as on land. This is deliberate: the purpose is to reject interpretations that put a survey point in the middle of the Atlantic, not to police whether something is a few hundred metres offshore. A coordinate on a beach, in a harbour, on a pier, or just off the coast will pass the land check and can still rank as a Strong match. This is correct behaviour, not a bug. If you expect your coordinate to be genuinely offshore (a buoy, a drilling platform, a survey point in open water), the tool will still identify the right system as long as the point is within the system's coverage area, even if the map preview shows it over water. The land check only downgrades a candidate when the point lands far enough offshore that the interpretation is implausible, which typically means open ocean well beyond any coastline.
Troubleshooting coordinate identification
- "No interpretation survived." Every tested system placed the pair at sea, outside its own coverage, or outside your hint. Remove the hint and try again, then check whether the values are in the expected order. If the numbers are very small or very large for a grid, they may be in a local or engineering coordinate system that no public tool can identify.
- The right answer is not in the list. This version tests British National Grid, UTM and US State Plane (NAD83). Other national grids such as Irish Grid, Swiss LV95, Lambert 93 or Gauss-Krüger are not yet included. Their values will often show up as UTM or State Plane guesses that land somewhere wrong, which is itself a useful signal.
- Too many UTM zones. Add a region hint. Without one, the numbers genuinely cannot distinguish zone 30 from zone 31.
- The hint made everything disappear. The hint is treated as a hard filter. If the data really is from elsewhere, or the pair is a lat/long in the wrong order that lands on another continent, clearing the hint will bring the candidates back.
- Only one number was found. A coordinate needs both values. Paste the easting and northing (or latitude and longitude) together, separated by a space or comma.
- Feet or metres? For State Plane the tool tries both. If both survive for the same zone they will appear as separate cards; the one that lands in a sensible place is the one to use.
Once you have picked an interpretation, Use this and convert sends the point into the map converter with that system already applied, where you can see it in every other format and change the datum. Have a whole file of the same kind? The bulk converter processes them all at once.